OpenCode Desktop Just Got WAY More Useful

[00:00:00]Open Code Desktop just got way more useful. What if you could run two AI coding helpers at the exact same time?

[00:00:07]Not one after the other, both. Right now, most people using Open Code have no idea this just dropped. And it makes the whole app feel brand new. Ready to see it? I'm the digital avatar of Julian Goldie, and I help you actually learn and use AI tools in your real work, not just watch demos and forget them. In this video, I'll show you the new update to Open Code Desktop, how I used it to build two things at once for the AI profit boardroom, and one free trick near the end that makes the whole app even better. Stick around because that last part changes how you work. First quick one, what even is Open Code? Open Code is a free open-source AI coding agent. Think of it as a smart helper that writes code for you, edits your files, and runs commands. You can use it in your terminal, in your code editor, or in a desktop app. And here's the part people love. You can plug in almost any AI model, Claude, GPT, Gemini, and loads more. There are even free models built right in, so you can start without setting anything up. It's used by millions of developers every month. So, this isn't some tiny side project. This is a real tool that a lot of people trust. Now, here's the big update. Open Code Desktop is now built around tabs.

[00:01:08]Yes, tabs. Just like the tabs in your web browser. Before this, you kind of lived in one window at a time. Now, you can open a tab, start a fresh session, and get to work. open another tab and start something totally different. Each tab holds its own session. You can jump between them whenever you want. And this works on Mac, Windows, and Linux. So, no matter what computer you're on, you get it. So, what does that actually give you? A few simple things that add up fast. One, you can start a new session in a tab anytime you begin something new. Two, you can open an old session from any of your projects right in a tab. So, your past work is never lost.

[00:01:42]It's just one click away. Three, when you're done with something, you just close the tab. Clean and simple. It sounds small, but if you've ever had five things going at once and kept losing your place, you'll feel the difference right away. Here's how it works in real life. You open the desktop app. You see your tabs across the top.

[00:01:57]You click to make a new one. You pick your project and your model. You type what you want. That's it. Each tab is its own little workspace. And because each tab runs its own session, one tab doesn't mess with another. They just run side by side like two workers at two desks. And here's something handy. Each tab can use a different model. So, one tab can run a free model while another tab runs a stronger one. You're not locked into one choice across the whole app. You pick per tab. This is huge when you want to test the same idea two ways.

[00:02:23]Run it with one model in this tab, run it with another model in that tab, then compare which one did the job better. No restarting, no losing your spot. You just click across the top and keep moving. That's the whole point of tabs.

[00:02:33]Let me show you what I mean because this is where it clicked for me. I opened Open Code Desktop and made two tabs. In the first tab, I picked one of the free models and I typed one line, make an HTML website for my AI profit boardroom community. Then, instead of sitting there waiting, I opened a second tab. In that second tab, I picked another free model and I asked it to build a simple onboarding checklist page for new AI profit boardroom members, a page that shows them the first steps to take on day one. Now, here's the wild part. Both tabs were working at the same time. One model was building the AI Profit Boardroom website. The other model was building the onboarding checklist for the AI profit boardroom right next to it. I wasn't waiting on one to finish before I started the other. I flipped between the two tabs and watched them both work and they both finish the job.

[00:03:18]Two builds, two models, one app at the same time. This is exactly the kind of thing I do for the AI profit boardroom.

[00:03:24]I'll spin up a landing page in one tab while a helper tool for AI profit boardroom members gets built in the other. And if I wanted a third thing, I'd just open a third tab and keep going. Okay, now the free trick I promised. You can use Open Code Desktop with Olma. If you've never heard of Olma, it's a tool that lets you run AI models right on your own computer, local on your machine. That means the model isn't sitting on someone else's server.

[00:03:46]It's running on your laptop. Why does that matter? A few reasons. It can be more private because your work stays on your computer. You're not sending every request out to an outside service and it keeps working even when your internet is being slow. Setting this up used to feel like a chore. Not anymore. Ola can now set up Open Code for you. You run one short command, Olma, launch Open Code.

[00:04:06]That's it. It gets Open Code ready to use a model right on your computer. One thing to know though, for coding work, you want a model with a big enough memory window. Open Code likes a context length of 64K or higher. That's just a fancy way of saying the model needs enough room to hold your project in its head. Pick a model that fits that and you're good to go. And picking your model is simple, too. Olma gives you a whole menu of models to choose from.

[00:04:28]Some run right on your machine and some run in the cloud if you want more power without a heavy download. For a lot of coding work, a solid local coding model does the job just fine. My advice, start small. Grab one model, get it working, and do a quick test in a tab. Once you see it run, then go bigger. Don't try to set up 10 models on day one. Get one win first, then build from there. So now you've got two big upgrades in one app.

[00:04:50]Tabs so you can run many things at once.

[00:04:52]An easy llama setup so you can run models right on your own machine. Two things that used to feel clunky, now smooth. So who actually needs this? Let me keep it real. If you only ever do one tiny task at a time, the old way was fine. But most of us don't work like that. If you're juggling a few projects, tabs are going to save your brain. If you like to keep your work on your own computer, the Olma setup is for you. If you're learning to build things and you want to try a few ideas fast, this is perfect. You can test one idea in one tab, another idea in the next tab, and compare them side by side. And the best part for a lot of people, Open Code is free, the free models are built in, and running local with Olmer keeps you in control of your own setup. This update takes an already strong tool and makes it feel easy. Now, if you're watching this and thinking, I want to actually set this up and not get stuck halfway, that's exactly what we do inside the profit boardroom. Inside the IP profit boardroom, I walk through Open Code step by step, the tabs, the model setup, and the Olma local setup all shown on screen so you can follow along at your own pace. We run live coaching calls where you can share your own screen and ask, "Hey, why won't my model connect or how do I set my context window?" And there's a road map that takes you from your very first tab all the way to running real projects without the guesswork. So if this update got you curious, the AI profit boardroom is where you go from curious to confident. Everything you're seeing me do here, we break it all down in there. Right, quick recap. So it six.

[00:06:13]Open code is a free open-source AI coding agent. The big update is Tabs on the desktop app, and it's on Mac, Windows, and Linux. Tabs let you run many sessions at once, open old ones from any project, and close them when you're done. I showed you two tabs building two things at the same time and you can now set up Olma in one short command to run models right on your own computer. Just remember that 64K context window and you're set. Before you go, three quick pro tips. Tip one, use one tab per job. Don't cram five ideas into one session. Give each idea its own tab so your thinking stays clean. This is how I keep my AI profit boardroom projects from getting tangled up. Tip two, make your first message in a tab clear and specific. When you've got a bunch of tabs open, a clear first line helps you spot the right one fast instead of clicking around lost. Tip three, start with a free model to test your idea, then switch to a stronger one for the final version. I do this all the time for AI profit boardroom builds. I rough it out first, then I polish it.

[00:07:10]Small habits like these save you a ton of time. If you want the full process, SOPs, and 100 plus AI use cases like this one, join the AI success lab. Links in the comments and description. You'll get all the video notes from there, plus access to our community of 85,000 members who are crushing it with AI. And if you're ready to really run with this, here's my last word. You're about to open Open Code, make your first tabs, and try to wire up a model. And somewhere in there, something won't connect or a model won't load, and you'll wish someone would just show you the fix. That's what the AI profit boardroom is for. Inside you get the open code tutorials, the live coaching calls to unstick you fast, the road map to follow step by step, and the exact prompts I use so you're never staring at a blank screen wondering what to type.

[00:07:54]With over 4,000 members building right alongside you, you won't be stuck for long. Come join us at AI profitboardroom.com.